Lab Science News - Science, chemistry and environmental news from laboratory experts
 

Vapor Intrusion Investigations: Air Sampling Tips for Meeting Data Quality Objectives

September 26th, 2011

N. Dagnillo1, L. Hill2, A. Fortune3, A. Smith4, and S. Thompson2
1Trihydro Corporation, 3001 E. Pershing Blvd, Suite 115, Cheyenne, WY 82007
2Trihydro Corporation, 1537 Riverside Ave., Suite 101, Fort Collins, CO 80524
3Columbia Analytical Services, Inc., 2655 Park Center Drive, Suite A, Simi Valley, CA 93065
4Trihydro Corporation, 9460 Calle Milano, Atascadero, CA 93422

Vapor intrusion is a fate and transport process characterized by the upward movement of volatile chemicals from subsurface contamination (e.g., buried waste, contaminated groundwater) into overlying buildings. The potential for adverse human health effects from exposure to indoor air vapors has motivated private, state, and federal entities to develop guidance documents and protocols specific to the collection and analysis of soil vapor data.

 

Generalized schematic of the pathway for subsurface vapor intrusion into indoor air.

Figure 1: Generalized schematic of the pathway for subsurface vapor intrusion into indoor air.

While the sample collection methodologies have improved over time, specifics related to laboratory quality control and data validation have not been comprehensively addressed. The United States Environmental Protection Agency (USEPA) Contract Laboratory Program National Functional Guidelines (NFG) for Organic and Inorganic Data Review (USEPA 2008 and 2010) were developed to provide practitioners consistency and accountability when evaluating and reviewing laboratory analytical data produced from soil and water samples. However, the USEPA has not yet published guidelines specific to the evaluation of soil vapor data. Documenting the validity of soil vapor data is a key component of conducting any vapor intrusion pathway assessment since many of these studies are conducted to support human health risk assessment. In the absence of applicable guidance, Trihydro Corporation has developed best practices from experience in field collection of soil vapor samples, validation of the resulting data, and careful study of soil vapor analytical methods.

The key components of these best practices are sample planning, sampling technique, laboratory analysis, and validation of the resulting analytical data. The following sections provide the components of this best practice approach.

Sample Planning

Sampling equipment cleanliness plays a crucial role in soil vapor sampling and data quality analysis. Since canisters and associated equipment (e.g., flow controllers, vacuum gauges, etc.) are reused, equipment cleaning and certification are essential to obtaining good data quality. If equipment is not properly cleaned, artifacts can remain in the equipment, resulting in a high bias. This phenomenon was quantitatively explored in a paper by Fortune, et. al. (2008). Laboratory experiments confirmed that when flow controllers and/or vacuum gauges were reused without adequate cleaning, carryover from high concentration samples resulted in false positives over the laboratory reporting limits.

After a soil vapor sample has been analyzed in the laboratory, the canister is subjected to a cleaning process using cycles of heat, humidification, and flushing with a clean gas source (nitrogen or air). Canisters are cleaned on a manifold in small batches of up to twenty. Flow controllers, vacuum gauges, and other components of the sampling train are often cleaned in a similar manner. To measure the effectiveness of the cleaning process, the equipment is analyzed for the compounds of interest. To achieve a clean certification, all compounds of interest should be below the laboratory detection limits.

Canisters and associated equipment may be either batch or individually (100%) certified as clean. Both options are compliant with Method TO-15; choosing which option to use will depend on the data quality objectives of the sampling program.

  • Batch Certification – One canister per cleaning batch is deemed representative of all canisters in the batch and analyzed for the compounds of interest. Many laboratories will further employ the best practice of certifying the canister that displayed the highest concentrations in its previous use. If the chosen canister passes certification, the entire batch is certified as clean. The consequence of batch certification is that the data user does not have complete certainty that each canister was clean before use in the field.
  • 100% Certification – Each canister is analyzed for the compounds of interest. Flow controllers, vacuum gauges, and other components of the sampling train may be certified in a similar manner. When using 100% certified canisters, it is important that the certified pieces (canister, flow controller and vacuum gauge) be used together and not mixed with other certified units. While significantly more labor intensive (and therefore more costly), this option gives the data user confidence that every piece of equipment used was certified as clean prior to sampling. Individual certification is often recommended at sites with sensitive receptors (e.g., schools, daycare facilities) and for sampling programs that may be involved in litigation.

Compounds likely to exhibit carryover in sampling equipment include toluene, tetrachloroethene, trimethylbenzenes, naphthalene, and other similar VOCs with molecular weights over approximately 160 g/mol and/or boiling points over approximately 160-170°C. Lighter molecular weight/boiling point compounds are usually more easily removed during the cleaning process. Excessively high concentrations (greater than approximately 100,000 μg/m3) of any compound may also have a tendency for carryover; therefore, equipment exposed to these concentrations may require additional cleaning.

Equipment may also be segregated by use and concentration level in the laboratory to reduce potential residual carryover in the sampling equipment and/or analytical system when low reporting limits are needed. For example, canisters used for ambient air and indoor sampling, which typically require low-level reporting limits, could be segregated from canisters used for sampling soil gas, which tend to have higher levels of contaminants.

Sampling Technique

Field processes can greatly impact the quality of sample data. To collect representative soil vapor samples, it is important to limit leakage across the sample train, which can consist of the vapor probe, flow controller, nylon plastic tubing, compression fittings, ball valves, and the sample canister. It is also important that any components used in the sampling train be evaluated to determine if they have the potential to cross contaminate the sample. The following quality control measures can be implemented when collecting soil vapor samples to improve data quality.

Canister and Sample Train Leaks

Canister and sample train leaks can allow ambient air to enter the sampling system, resulting in the collection and analysis of non-representative samples. Common causes of leaks include worn or faulty valves and fittings, over-tightened fittings, improperly installed vapor probes, and broken or cracked tubing.

Valves and connection fittings will wear over time. If a valve or fitting is used for more than one sample, it should be thoroughly decontaminated and inspected for wear. The threads should be intact, and valves should close completely. Stainless steel medical grade valves and fittings are recommended since they are harder than brass and tend to be more durable. Fittings should be snug, but not over tightened. Over tightening fittings will cause damage to the threads and can also result in crimping of the tubing. Crimping may create a gap in the connection, allowing ambient air to enter the system. Furthermore, nut and ferrule components should also be comprised of stainless steel, as plastic components are more susceptible to warping during tightening, allowing gaps in the connection.

Nylon plastic tubing, which is commonly used for vapor sample collection, is moderately flexible and usually works well with the typical nut and ferrule fittings. Tubing should be used only once to prevent cross contamination. Nylon plastic tubing becomes brittle in cold weather and may crack or break. In order to avoid cracking and breaking, tubing should be inspected prior to connecting to the sample train, and sections should be cut to lengths that will accommodate bending and placement of the sample train components.

Proper installation of a vapor probe, whether a sub-slab or soil vapor sampling point, is critical to collecting high-quality samples. Sub-slab vapor probes should be properly sealed to the surface of the slab using a non-volatile material such as quick-setting hydrated cement. The seal around the vapor probe should be allowed to dry thoroughly prior to attaching components of the sample train. Soil vapor probes should be installed and sealed using impermeable barriers such as bentonite. For nested soil vapor wells, a seal test should be conducted by applying a vacuum to a vapor probe while monitoring the next deeper or shallower probe with a vacuum gauge. A measurable pressure change may indicate an improper seal and require abandoning the sampling point.

Canister Pressure Monitoring

Canisters are prepared in the laboratory and are typically shipped to sampling locations with a pressure as close to 30.0 inches of mercury (inHg) as possible, depending on the laboratory’s elevation. Atmospheric pressure can range from 29.2 inHg at sea level to 24.9 inHg at 5000 feet. Once a canister is received at the sampling location, the canister pressure should be measured by the sampling team and compared to the laboratory measurement recorded on the receiving documents. Field and laboratory canister pressure readings should be consistent, accounting for differences in pressure due to the change in elevation from the laboratory to the sampling site, and differences in the gauge between the laboratory and the field. The MassDEP APH Method suggests an acceptable difference of plus or minus 5 inHg (MassDEP 2010). However a site specific value could be determined. Use of a digital gauge is recommended over the gauges provided by the laboratory, as the laboratory gauges tend to be less accurate due to wear and tear during shipment. Canisters exhibiting an excessive loss of vacuum from the laboratory to the sampling location should be considered to have leaked in transport and should not be used for sampling.

In addition to monitoring the canister pressure from the laboratory to the sampling location, the pressure should be monitored and recorded after sampling and prior to transport to the laboratory. The laboratory will measure the pressure again upon receipt and report this reading along with the sample results. A residual vacuum should remain in the canister after sampling. This residual vacuum allows the data user to determine if canister leaks have occurred during shipment from the site to the laboratory following sample collection. Canisters exhibiting an excessive loss of vacuum from the sampling location to the laboratory to the sampling location should be considered to have leaked in transport, and the resulting data should be rejected.

Shut-In Testing

Shut-in testing is a technique used in the field to evaluate the integrity of the sample train. A typical shut-in test involves closing the valve to the vapor probe and evacuating the lines to a measured vacuum. Then, the vacuum is shut by closing the valve at the opposite end of the sampling train. This vacuum is monitored using an in-line vacuum gauge. The initial measured vacuum should be maintained for 30 seconds for a positive shut-in test. If the vacuum dissipates from the sample train, the shut-in test has failed, and all connections should be inspected for leaks.

Purging and Tracer Testing

Purging, along with the use of a tracer compound, is performed to remove ambient air from the vapor probe and tubing and to check for leaks in the sample train. Tracer compounds can include gaseous materials such as helium and 1,1 diflouroethane (duster spray) and liquid compounds such as isopropyl alcohol, pentane, and Freons. During purging, the sample train is shrouded, and a tracer compound is applied within the shroud. Soil vapor is then purged from the vapor point into a Tedlar bag, where it can be monitored using field instruments.

There are advantages and disadvantages to all tracer compounds (ITRC 2007). Selection of a tracer is specific to the project needs but should include consideration of:

  • Accessibility of the tracer
  • Toxicity
  • Interference with analytical results
  • Real-time monitoring
  • Quantitative evaluation of leakage

Since liquid tracer compounds are typically measured in the laboratory (after sampling is complete), the use of helium is prevalent in the field because it can be monitored using a hand-held helium detector. Helium is non toxic and can be purchased from gas suppliers. Ultra high purity grade helium should be used to limit the potential for VOC cross contamination that can occur with party-grade helium. Real-time monitoring of helium in the field allows the sampler to determine if the tracer gas has infiltrated the sample train. The concentration of helium measured in the Tedlar bag should read less than 5% of the average shroud concentration. If the helium measurement is greater than 5% of the average shroud concentration, the sample train should be inspected for leaks and retested. Should leakage occur in the field, the percent can be quantified using the laboratory results along with the known shroud concentrations measured in the field during sample collection. This allows the data user to determine if the leak was significant enough to warrant the data unusable. Other tracers that do not allow for field measurement can only be qualitatively evaluated, thereby resulting in less defensible data.

Additionally, the use of helium as a tracer is recommended since its presence will not interfere analytically with the Method TO-15 VOC analysis. For example (as discussed in ITRC 2007), if the sampling system had a 0.1% leak (i.e., below 5%, and considered acceptable by any regulatory agency) and isopropyl alcohol was used as a tracer, the resulting concentration of isopropyl alcohol in the Method TO-15 sample (making conservative assumptions) would be over 14,000 μg/m3. This tracer concentration would force the lab to perform a dilution on the sample, thus raising reporting limits of all target VOCs.

Laboratory Analysis

When choosing a volatile organic compounds (VOC) analytical method for soil vapor sample analysis, a few common options exist: USEPA Method TO-15, which utilizes gas chromatography/mass spectrometry (GC/MS) and was originally written for ambient air samples; and USEPA Test Methods for Evaluating Solid Waste, Physical/Chemical Methods (also known as SW846) Method 8260B, which utilizes GC/MS and is widely used for groundwater and solid waste samples.

While Method 8260B technically allows for analysis of vapor phase samples (“air” is mentioned in Section 1.2), the method does not precisely define a procedure for vapor phase sample introduction and/or calibration considerations with typical 8260B purge and trap analytical equipment; therefore, analytical procedures must be significantly modified by the laboratory to accommodate vapor phase samples. Because of this ambiguity, there is no unified approach among laboratories for introducing soil vapor into the analytical system (Tuday 2008). Adding to the inconsistency, state certifying agencies do not typically certify laboratories (fixed or mobile) specifically for soil vapor analysis via 8260B. Potential losses of vapor phase samples in the analytical system should be monitored via additional quality control such as a vapor phase check standard. This protocol (called a tertiary standard) has been adopted by the Arizona Department of Environmental Quality (ADEQ 2009). The tertiary standard is comprised of a third source vapor standard and is used to verify the recovery of the vapor standard compared to the purged aqueous standard. Additional quality control is performed on the purge liquid (e.g., surrogates, internal standards, matrix spikes, etc.); this additional tertiary standard, if performed, is the only quality control directly relevant to the vapor phase.

Method TO-15 is a standardized method specific to the vapor phase. Although Method 8260B can successfully be used as an investigative tool for certain applications (if performed with the appropriate sample introduction and quality control), Method TO-15 is preferred when soil vapor data are intended for use in support human health risk assessments. Various laboratory accreditation authorities routinely audit and certify laboratories for this method, helping to ensure inter-laboratory consistency. All quality control parameters in Method TO-15 are specific to the vapor phase (calibration standards are in the vapor phase, control samples and method blanks are in the vapor phase, etc.).

Other canister-based USEPA methods are occasionally referenced in guidance documents, quality assurance project plans (QAPPs), and/or work plans. USEPA Method TO-14 was originally published in March 1989 in the Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air. In January 1999, Method TO-14 was revised and updated as Method TO-14A in the Second Edition of the Compendium of Methods; as such, Method TO-14 has been superseded by Method TO-14A. Method TO-15 was a new method added to the Second Edition of the Compendium in January 1999. Method TO-15 is larger in scope and better defined for the analysis of VOCs in air and other gaseous matrices than Method TO-14A. In practice, Method TO-15 has supplanted Method TO-14A as the preferred method for the analysis of VOCs in air.

Method TO-15 can be used to quantify a subset of individual VOCs, typically from C3 to C12. A vapor phase calibration standard is analyzed for each compound of interest, and an internal standard calibration technique is used. Although originally written for ambient air, Method TO-15 can be applied to the soil vapor matrix. Since no specific compound list is specified in the method, the compound list of VOCs may vary from laboratory to laboratory; however, most laboratories will report concentrations of common VOCs such as benzene, toluene, ethylbenzene, xylenes, tetrachloroethene, trichloroethene, and other chlorinated solvents.

When analyzing a sample via Method TO-15, the data user may request that the laboratory also perform (on the same sample) an analysis of “Tentatively Identified Compounds (TIC).” The TIC analysis will identify and/or give an estimated concentration for any compound present that is not otherwise reported in the Method TO-15 analysis. Since adding the TIC analysis will give a more complete list of VOCs in the sample, this additional step is often useful if the full characterization of a site is unknown.

For petroleum hydrocarbon sites, analyzing a canister sample via the Massachusetts Department of Environmental Protection (MassDEP) Air-Phase Petroleum Hydrocarbon (APH) Method (MassDEP 2010) may be useful. This method, based on Method TO-15, reports target VOCs associated with petroleum products, but it also reports aliphatic and aromatic hydrocarbon fractions (C5-C8 aliphatic hydrocarbons, C9-C12 aliphatic hydrocarbons, C9-C10 aromatic hydrocarbons) for risk assessment purposes. Non petroleum-related compounds that fall between the retention time markers of a hydrocarbon range (e.g., d-limonene, a common essential oil found in household cleaners, within the C9-C12 aliphatic range) may be subtracted per the data user’s request. The results will render a more accurate representation of the compounds related to petroleum product. While other similar fractionated approaches exist for soil and groundwater analyses in other states (e.g., the Northwest Total Petroleum Hydrocarbons (NWTPH) methods), the Massachusetts APH Method is the only one of its kind to specifically address petroleum product in the vapor phase. Both Method TO-15 and the Massachusetts APH Method can be performed on the same canister sample.

Other laboratory analyses that may be performed on soil vapor samples include an analysis for a leak test tracer (e.g., helium) and an analysis for permanent gases associated with aerobic biodegradation (methane, carbon dioxide, oxygen) via USEPA Method 3C or American Society for Testing and Materials (ASTM) Method D1946. In most cases, all of the analyses may be performed out of the same canister. In some cases, a separate canister or Tedlar bag sample may be required; the data user should coordinate in advance with the analytical laboratory.

Data Validation

Analytical data quality is an important aspect of environmental investigations. Understanding the different components that may affect data quality can provide a greater understanding of the results and the possible shortfalls of the data. The USEPA Contract Laboratory Program NFGs for Superfund Organic Methods Data Review provides guidance for validating analytical data quality; however, this specific guidance includes data produced from solid and water matrices and methods specifically applicable to those matrices. Analytical air methods, such as Method TO-15, document sampling and analytical procedures for the measurement of some VOCs and some hazardous air pollutants; however, the methods provide procedures under conditions typical of those encountered in routine ambient air analysis. The methods do not account for all possible conditions. Therefore, data validation best practices are used to provide valuable data with respect to the multiple elements that may affect sample quality. Using Method TO-15 and the NFG requirements as a basis, best practices for data validation of soil vapor data are presented below.

Sample Dilutions

Assessing the validity of sample data involves reviewing sample dilution factors to verify that they are reasonable based on the canister pressures upon receipt at the laboratory. Traditionally, analytical samples require dilutions for the following two reasons:

  • Sample dilution can be the result of the addition of humidified zero air to the sample upon receipt at the laboratory to achieve a positive pressure in the canister, or a pressure that is slightly above ambient pressure. Typical positive pressure amounts are around 5 pounds per square inch (psi); however, no specific requirement exists. Laboratories often use the reasoning that older instruments perform better with canisters under a slight positive pressure, and some studies have shown that VOC analytes are more stable under a slight positive pressure (Coutant 1992). In addition, pressurization of canisters allows for more accurate sample loading, provides sufficient volume for reanalysis/dilutions, and reduces the chance of cross contamination of autosampler equipment. The amount of humidified zero air introduced to the sample canister is directly related to the amount of sample volume collected in the field. The more sample volume collected within the sample canister, the less humidified zero air that is added, resulting in less sample dilution. Therefore, the sampler should attempt to provide enough sample volume so that approximately 2 to 5 inHg remains in the canister at the end of sampling. As discussed previously, some residual vacuum should remain in the canister to allow for determination of leakage during transit from the field to the laboratory.
  • A sample dilution can be the result of the laboratory analyzing a smaller sample volume. The decision to analyze a smaller sample size is usually the result of a highly contaminated sample or of compounds outside of laboratory calibration ranges.

A review of analytical data dilution factors can reveal several reasons for sample dilutions. However, when assessing the sample dilutions, the data user should discuss with the laboratory the reasoning behind the addition of makeup gas. In addition, the data user should speak with the laboratory about reporting limit requirements prior to sample collection. The conversation will allow the laboratory to provide information regarding the sample volume required to meet specific reporting limit requirements.

Instrumentation Quality Control

Depending on the method used, instrument quality control assessments may include some or all of the following: initial and continuing calibrations, instrument performance check results, and internal standard results.

Evaluating Calibration Results

Analytical calibration results are assessed to monitor the instrument’s initial and continued ability to successfully produce qualitative and quantitative data according to the analyte list. Method TO-15 defines dynamic calibration as “calibration of an analytical system using calibration gas standard concentrations in a form identical or very similar to the samples to be analyzed and by introducing such standards into the inlet of the sampling or analytical system from a manifold through which the gas standards are flowing.”

Instrument calibrations for air samples are analyzed in two parts. The first part is an initial calibration that consists of at least five concentrations of the target analytes, which span the monitoring range of interest. Second, the daily continuing calibrations are prepared and analyzed from the midpoint standard used in the initial calibration sequence. The continuing calibration verifies that the instrument has maintained the appropriate sensitivity and linearity established in the initial calibration. Assessments of calibration results are based on several of the following components (USEPA 2008).

Relative Response Factor (RRF) and Mean RRF: The RRF is a measure of the relative response of the instrument detector of an analyte compared to an internal or external standard. The RRFs are determined by the analysis of standards and are used to calculate the concentrations of analytes in samples. Since specific guidelines are not provided in Method TO-15, the NFG limits are used to evaluate the RRF and Mean RRF values. According to the NFG, the RRF and mean RRF values should be equal to or greater than 0.050, with the exception of the analytes noted in Table 15 of the NFG. The RRF values for these compounds should be equal to or greater than 0.01 (USEPA 2008).

Percent Relative Standard Deviation (%RSD): The %RSD is assessed using the initial RRF values. The %RSD is a measure of how precise the average is and how well the individual numbers agree with each other. Per the NFGs, most analytes must produce a %RSD less than 20%. However, Method TO-15 specifies that the %RSD be less than 30%. When validating air data, it is recommended that the %RSD limit of less than 30% be used since it is more difficult to produce precise results between air samples (as compared to water samples.). Additionally, the vapor phase standards used in method TO-15 are prepared with a higher error tolerance than the liquid phase standards used in analyses typically validated using the NFGs. For example, a vapor phase standard may have an error of approximately plus or minus 10% and a liquid standard may have an error of plus or minus 5%. Analytical results associated with continuing calibration %RSD values greater than 30% are qualified as J or UJ to indicate estimated concentrations or reporting limits.

Percent Difference (%D): The %D is assessed using the continuing calibration results. The %D is calculated between the initial calibration mean RRF and the continuing calibration RRF. Per the NFGs, most analytes must produce a %D that is less than 25%, with the exception of the analytes listed in Table 15, which must produce a %D that is less than 40%. However, Method TO-15 specifies that the %D be below 30%. When validating air data, it is recommended that the %D limit of less than 30% be used, with the exception of the Table 15 analytes, which should use a %D limit of less than 40%, since it is more difficult to produce precise results between air samples (as compared to water or soil samples).

When assessing the validity of air data, the procedures in Table 1, below, are recommended. The validator should verify that the frequency of calibration is in accordance with the method that is being used. (Method TO-15 requires that calibration verifications be analyzed every 24 hours.)

Table 1: Criteria for Reviewing Calibration Results

Criteria for Analysis*

Flagging Criteria**

Detected Associated Compounds Non-Detected Associated Compounds
RRF < 0.050 (or < 0.010 if listed in Table 15 of the NFG)

J or R (based on mass spectral identification)

R

RRF ≥ 0.050 (or > 0.010 if listed in Table 15 of the NFG)

No Qualification

%RSD > 30% (or > 40% if listed in Table 15 of the NFG)

J

UJ

%RSD ≤ 30% (or < 40% if listed in Table 15 of the NFG)

No Qualification

The J (for a detected value) and UJ (for an undetected value) flags indicate that the concentrations and contract required reporting limits are estimated. The estimated results may be inaccurate or imprecise, but are considered acceptable for use during a site assessment.

The R flag indicates that the results are rejected and should not be used during site assessment.

*Criteria for analysis was derived from the USEPA NFGs for Superfund Organic Methods Data Review and Method TO-15

**Flagging criteria were intended to model those established in the USEPA NFGs for Superfund Organic Methods Data Review

Evaluating Instrument Performance Check Results

For VOC analyses, the GC/MS instrument must meet tuning and standard mass spectral abundance criteria prior to initiating data analysis. The GC/MS system is set up according to the manufacturer’s specifications, and the mass calibration and resolution of the GC/MS system are then verified by the analysis of the instrument performance check standard, bromofluorobenzene (BFB) (USEPA 1999). In evaluating the instrument performance check data, the frequency of analyses should be evaluated. The instrument performance check standard should be analyzed before calibration standards and then as required by the analytical method. Method TO-15 requires the analysis and evaluation of the mass spectrum of BFB every 24 hours following the initial run.

Daily calibration sequences for the analysis of air samples begin with the injection of BFB. The calibration standard may be analyzed only if the BFB mass spectrum meets the ion abundance criteria. Table 2, below, includes important Method TO-15 evaluation criteria for BFB. When assessing the validity of data using the instrument tune results, the validator should reject results reported with an instrument tune outside of the criteria noted in Table 2, below, or those reported in the method employed.

Table 2: Criteria for Instrument Tune Results Mass Ion Abundance Criteria

Mass

Ion Abundance Criteria*

50

8.0 to 40.0% of m/e 95

75

30.0 to 66.0% of m/e 95

95

Base Peak, 100% Relative Abundance

96

5.0 to 9.0% of m/e 95

173

Less than 2.0% of m/e 174

174

50.0 to 120.0% of m/e 95

175

4.0 to 9.0% of m/e 174

176

93.0 to 101.0% of m/e 174

177

5.0 to 9.0% of m/e 176

*All ion abundances must be normalized to m/z 95, the nominal base peak, even though the ion abundance of m/z 174 may be up to 120% that of m/z 95.

Evaluating Internal Standard Results

Internal standard performance criteria are designed to ensure that GC/MS sensitivity and response are stable during each analysis (USEPA 2008). Internal standard spikes are injected with each field and quality control sample. Assessing the validity of internal standard results involves verifying that each sample was spiked with the applicable internal standard and assessing the area count results and retention times of the internal standards. Method TO-15 indicates that the area count of the internal standards should be within 40% of the mean area response over the initial calibration range. Additionally, Method TO-15 indicates that the retention times of the internal standards should not vary more than plus or minus 20 seconds of the mean retention time over the initial calibration range. Table 3, below, provides criteria for assessing internal standards.

Table 3: Criteria for Reviewing Internal Standard Results Criteria

Criteria*

Flagging Criteria*

Detected Associated Compounds

Non-Detected Associated Compounds

Area response is within 40% of the mean area response of the internal standard over the initial calibration range

No Qualification

Area response is > 140% of the mean area response of the internal standard in the most recent valid calibration

J

No Qualification

Area response is < 60% of the mean area response of the internal standard in the most recent valid calibration

J

R

Retention time for the internal standard of the sample is within ±20 seconds of the most recent valid calibration internal standard.

No Qualification

Retention time for the internal standard of the sample is not within ±30 seconds of the most recent valid calibration internal standard.

J

R

The J (for a detected value) and UJ (for an undetected value) flags indicate that the concentrations and reporting limits are estimated.  The estimated results may be inaccurate or imprecise, but are considered acceptable for use during a site assessment.  

The R flag indicates that the results are rejected and should not be used during site assessment.

*Flagging criteria are intended to model those established in the USEPA NFGs for Superfund Organic Methods Data Review

Method Blank and Equipment Blank Samples

The purpose of blank analyses is to determine the existence and magnitude of contamination resulting from laboratory or field activities. The criteria for evaluation of blanks apply to any blank associated with the samples. If problems with any blank exist, all associated data must be carefully evaluated to determine whether or not there is an inherent variability in the data or if the problem is an isolated occurrence not affecting other data (USEPA 2008). When using 100% certified canisters, it is not necessary to collect trip blank or field blank samples.

Laboratory blanks can provide critical information on the validity of sample data and the source of contaminants and biases. Most analytical methods contain the following guidance: method blanks are analyzed with each analytical batch on a 1 per 20 sample basis or 1 per 24-hour period to evaluate possible contamination stemming from laboratory sources.

Equipment blank samples can be collected using helium passed through tubing or other equipment used in the field sampling process to check for contamination introduced by field sampling equipment.

Laboratory blank and equipment blank samples are analyzed identically to field samples. Table 4, below, can be used to determine if sample concentrations are authentic or possibly biased due to contamination from laboratory or field sampling equipment sources.

Table 4: Criteria for Reviewing Method Blank and Equipment Blank Samples

Blank Type

Blank Result

Sample Result

Action for Samples

Method or Equipment

Detects

Not Detected

No Qualification

< CRQL

< CRQL

Report CRQL value with a U

≥ CRQL

Use Professional Judgment

> CRQL

< CRQL

Report CRQL value with a U

≥ CRQL and < blank concentration

Report the blank concentration for the sample with a U or qualify the data as unusable R

≥ CRQL and ≥ blank concentration

Use Professional Judgment

= CRQL

< CRQL

Report CRQL value with a U

≥ CRQL

Use Professional Judgment

Gross Contamination

Detects

Qualify results as unusable R

The U flag indicates that the analyte was estimated to be undetected at the contract required reporting limit.

The R flag indicates that the results are rejected and should not be used during site assessment.

The table above was derived from the USEPA Contract Laboratory Program NFGs for Superfund Organic Methods Data Review Table 18, with some deviations.  Common laboratory contaminants are not held to a different standard since they are not expected to be a source of laboratory contamination in a laboratory that analyzes air samples.  

CRQL: Contract Required Quantitation Limit

Accuracy

Laboratory accuracy is a measure of system bias and can be measured by evaluating laboratory control samples (LCSs) and deuterated monitoring compounds (DMC or surrogate) percent recoveries and oxygen results. Laboratory Control Samples The LCSs are analyzed to assess laboratory accuracy using a blank sample with known concentrations of prepared standards. The accuracy of the analytical data is measured by the LCS percent recovery.

Equation 1: Percent Recovery

where:

Cs = Measured concentration of the spiked sample aliquot

Cu = Measured concentration of the unspiked sample aliquot (use 0 for the LCS or surrogate)

Cn = Nominal (theoretical) concentration increase that results from spiking the sample, or the nominal concentration of the spiked aliquot (for LCS or surrogate)

While the analytical methods recommend limits for the laboratory control sample percent recoveries, laboratories often provide limits based on statistical analysis of samples analyzed by the laboratory over a set period of time in their results. When assessing the validity of analytical data, the laboratory’s statistically based limits should be used. In addition, the data qualification criteria in Table 5, below, should be considered when assessing the validity of LCS data.

Table 5: Criteria for Validating Laboratory Control Samples Criteria Flagging Criteria

Criteria

Flagging Criteria*

Detected Associated Compounds

Non-Detected Associated Compounds

Percent Recovery is within the Laboratory Quality Control Limits

No Qualification

Percent Recovery is > than the Upper Laboratory Quality Control Limits

J

No Qualification

Percent Recovery is < than the Lower Laboratory Quality Control Limits

J

UJ

Percent Recovery is < than the Lower Laboratory Quality Control Limits and the Percent Recovery is Grossly Low.

J

R

The J (for a detected value) and UJ (for an undetected value) flags indicate that the concentrations and contract required reporting limits are estimated.  The estimated results may be inaccurate or imprecise, but are considered acceptable for use during a site assessment. 

The R flag indicates that the results are rejected and should not be used during site assessment.

*Flagging criteria are intended to model those established in the USEPA NFGs for Superfund Organic Methods Data Review.

Deuterated Monitoring Compounds

Laboratory performance on individual samples can be established by means of spiking the samples with DMCs. The sample itself may produce such factors as interferences (USEPA 2008). When DMCs are used in the analysis of samples collected in canisters, they cannot be considered to be true surrogates, as a true surrogate would follow a sample through the entire sample preparation. Since an air sample is not prepared (i.e., it is simply withdrawn from the canister, trapped and injected into the GC/MS system for analysis), and since the DMC compounds are added to the sample after the sample is withdrawn from the canister, DMCs for air samples do not necessarily have the same implications as DMCs for water or soil samples. Although none of the methods mentioned in this document require or mention the use of DMCs, USEPA Region 9 requires the use of DMCs for Method TO-15 (USEPA Region 9 2000). When DMCs are used, laboratories often provide limits based on statistical analysis of samples analyzed by the laboratory over a set period of time in their results. When assessing the validity of the analytical data, the laboratory’s statistically based limits should be used. In addition, the data qualification criteria in the Table 6, below, should be considered when assessing the validity of DMC recoveries.

Table 6: Criteria for Reviewing Deuterated Monitoring Compounds Recoveries Criteria Action

Criteria

Action*

Detected Associated Compounds

Non-Detected Associated Compounds

Percent Recovery is within the Laboratory Quality Control Limits

No Qualification

Percent Recovery is > than the Upper Laboratory Quality Control Limits

J

No Qualification

Percent Recovery is < than the Lower Laboratory Quality Control Limits

J

UJ

Percent Recovery is < than the Lower Laboratory Quality Control Limits and the Percent Recovery < 20%*

J

R

The J (for a detected value) and UJ (for an undetected value) flags indicate that the concentrations and contract required reporting limits are estimated.  The estimated results may be inaccurate or imprecise, but are considered acceptable for use during a site assessment. 

The R flag indicates that the results are rejected and should not be used during site assessment.

*Flagging criteria are intended to model those established in the USEPA NFGs for Superfund Organic Methods Data Review.

Assessing High Oxygen Results

Soil gas samples collected in canisters are often analyzed for VOCs, fixed gas analytes, and helium. Using one canister for all of these analyses may reveal certain problems with the fixed gas normalization calculations, particularly if helium has been used as a tracer. Historically, helium has been used by laboratories to pressurize canisters when oxygen is a target compound. However, when helium has been used as the tracer, the laboratory must use a different gas, typically nitrogen, to pressurize the canisters. The difficulties are illuminated when the canisters are pressurized with nitrogen upon arrival to the laboratory, since one of the analytes assessed in the fixed gas results is nitrogen. Following analyses, results for fixed gas analyses are normalized to 100%; therefore, the addition of nitrogen will increase that particular component and may cause other components to be high or low. When evaluating the validity and quality of fixed gas analyses, the following questions should be considered during the evaluation:

  • What type of gas was used to pressurize the canisters?
  • Can fixed gas samples be collected in a Tedlar bag instead of the canister used for VOCs and helium?
  • Are oxygen results around 21%? These questions and associated concerns should be discussed with laboratory personnel before and during data evaluation procedures. Oxygen results that are just above 21% may be explained by coelution with argon, but results that are much higher may require qualification or rejection of data.

Precision

Precision is the measure of variability of individual sample measurements. Field precision is determined by comparison of field duplicate sample results. Laboratory precision is determined by examination of laboratory duplicate results. Evaluation of field and laboratory duplicates for precision is performed using the %RPD, defined as the difference between two duplicate samples divided by the mean and expressed as a percent.

Equation 2:  Relative Percent Difference

C1 and C2 are the concentrations of duplicate samples.

Field Duplicates

A field duplicate sample is collected simultaneously with the primary sample at one sampling location. The same sample collection techniques are used for both samples. When assessing data validity using field and laboratory duplicate results, Table 7, below, may be used.

Table 7: Criteria for Reviewing Field Duplicate Sample Results Criteria

Criteria*

Flagging Criteria*

Detected Associated Compounds

Non-Detected Associated Compounds

The RPD is within the limits of 0 and 25%

No Qualification

The RPD is > 25%

J in the parent and duplicate samples

Not Applicable

The RPD could not be calculated since the compound was only detected in either the parent or duplicate sample.  However, the detected concentration was ≤ 2 times the reporting limit.

No Qualification

The RPD could not be calculated since the compound was only detected in either the parent or duplicate sample.  However, the detected concentration was > 2 times the reporting limit.

J in the parent or duplicate sample

UJ in the parent or duplicate sample

The J (for a detected value) and UJ (for an undetected value) flags indicate that the concentrations and contract required reporting limits are estimated.  The estimated results may be inaccurate or imprecise, but are considered acceptable for use during a site assessment. 

*Flagging criteria are intended to model those established in the USEPA Region 1 Revised Data Validation Guidance (USEPA Region 1, 1996), while taking into account the replicate precision requirement of 25% noted in Method TO-15.

Laboratory Duplicates

A laboratory duplicate is determined from the analysis of two samples prepared from the same canister. Laboratories often provide limits for laboratory duplicates in their results that are based on statistical analysis of samples analyzed by the laboratory over a set period of time. When assessing the validity of the analytical data, the laboratory’s statistically based limits should be used. In addition, the data qualification criteria in the Table 8, below, should be considered when assessing the validity of laboratory duplicate RPDs.

Table 8: Criteria for Reviewing Laboratory Duplicate Sample Results

Criteria*

Flagging Criteria*

Detected Associated Compounds

Non-Detected Associated Compounds

The RPD is < the laboratory quality control limits

No Qualification

The RPD is > the laboratory quality control limit

J

Not Applicable

The RPD could not be calculated since the compound was only detected in either the parent or duplicate sample.  However, the detected concentration was ≤ 2 times the reporting limit.

No Qualification

The RPD could not be calculated since the compound was only detected in either the parent or duplicate sample.  However, the detected concentration was > 2 times the reporting limit.

J

UJ

The J (for a detected value) and UJ (for an undetected value) flags indicate that the concentrations and contract required reporting limits are estimated.  The estimated results may be inaccurate or imprecise, but are considered acceptable for use during a site assessment. 

*Flagging criteria are intended to model those established in the USEPA Region 1 Revised Data Validation Guidance (USEPA Region 1, 1996).

Conclusions

The quality of soil vapor data is dependent on many factors, including media preparation, sampling technique, laboratory analysis, and validation of analytical results. In many cases, practitioners are being asked to provide data with very low reporting limits to allow for human health risk assessment of the vapor intrusion pathway. However, unlike soil and water analysis, the USEPA has not provided NFGs that are specific to the analysis of air. The aforementioned best practices establish a consistent approach for both improving, as well as validating, air data by providing specific guidelines for field methodologies and data validation approaches. Continued improvement in the evaluation and validation of air data will provide practitioners with defensible data that can be used to conduct vapor intrusion pathway assessments.

References

  1. Advanced Global Atmospheric Gases Experiment. Available from:
    http://agage.eas.gatech.edu/index.htm.
  2. Arizona Department of Environmental Quality (ADEQ). 2009. VOCs in Vapor by 8260B AZ Method, Revision 0.0, 04/14/2009. Available from:
    http://www.azdhs.gov/lab/license/tech/infoup.htm
  3. Coutant, R. W. 1992. Theoretical Evaluation of Stability of Volatile Organic Chemicals and Polar Volatile Organic Chemicals in Canisters. Battelle.
  4. Fortune, A. and Tuday, M. 2008. The Importance of Air Sampling Media Cleanliness for Vapor Intrusion Investigations. Presentation from National Environmental Monitoring Conference (NEMC), August 10-16, 2008, Washington, DC.
  5. Interstate Technology and Regulatory Council. 2007. Technical and Regulatory Guidance Vapor Intrusion Pathway: A Practical Guide. Available from:
    http://www.itrcweb.org/guidancedocument.asp?TID=49
  6. Massachusetts Department of Environmental Protection Bureau of Waste Site Cleanup (MassDEP). 2010. Quality Control Requirements and Petroleum Standards for the Analysis of Air-Phase Petroleum Hydrocarbons (APH) by Gas Chromatography/Mass Spectrometry (GC/MS) in Support of Response Actions under the Massachusetts Contingency Plan (MCP). Available from: http://www.mass.gov/dep/cleanup/laws/aphmcp.pdf
  7. Minnesota Pollution Control Agency. 2008. Risk-Based Guidance for the Vapor Intrusion Pathway. Available from:
    http://www.pca.state.mn.us/index.php/component/option,com_docman/task,doc_view/gid,3162
  8. Tuday 2008. Observations and Practical Experience in Soil Gas Analysis from a Laboratory Perspective. Presentation at the CA DTSC Soil Gas Advisory Forums, 2008. Presentation available at: http://www.dtsc.ca.gov/AssessingRisk/
  9. United States Environmental Protection Agency. 1989. Compendium Method TO-14, Determination of Volatile Organic Compounds (VOCs) in Ambient Air Using Specially-Prepared Canisters with Subsequent Analysis by Gas Chromatography. No reference exists, superseded by Method TO-14A.
  10. United States Environmental Protection Agency, Region 1. 1996. Revised Data Validation Guidance. Available from: http://www.epa.gov/region1/oeme/index.html
  11. United States Environmental Protection Agency. 1999. Compendium Method TO-15, Determination of Volatile Organic Compounds (VOCs) in Air Collected in Specially-Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (EPA/625/R-96/010b). Available from: http://www.epa.gov/ttnamti1/files/ambient/airtox/to-15r.pdf
  12. United States Environmental Protection Agency. 1999. Compendium Method TO-14A, Determination of Volatile Organic Compounds (VOCs) in Ambient Air Using Specially-Prepared Canisters with Subsequent Analysis by Gas Chromatography (EPA/625/R-96/010b). Available from: http://www.epa.gov/ttnamti1/files/ambient/airtox/to-14ar.pdf
  13. United States Environmental Protection Agency, Region 9. 2000. Volatile Organic Compounds (VOCs) in Air (Ambient Air/Soil Vapor/Stack Gas) Samples Collected in Specially-Prepared Canisters and Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS). Available from: http://www.epa.gov/region9/qa/pdfs/dqi/vocs_gc.pdf
  14. United States Environmental Protection Agency. 2002. OSWER Draft Guidance for Evaluating the Vapor Intrusion to Indoor Air Pathway from Groundwater and Soils (Subsurface Vapor Intrusion Guidance) (EPA530-D-02-004), Appendix E. Available from:
    http://www.epa.gov/osw/hazard/correctiveaction/eis/vapor/complete.pdf
  15. United States Environmental Protection Agency. 2008. USEPA Contract Laboratory Program National Functional Guidelines for Superfund Organic Methods Data Review (USEPA-540-R-08-01). Available from: http://www.epa.gov/superfund/programs/clp/download/somnfg.pdf
  16. United States Environmental Protection Agency. 2009. USEPA Region 2 SOP HW-31 Revision 4 Validating Volatile Organic Analysis of Ambient Air in canister by Method TO-15 (SOP # HW-31). Available from: http://www.epa.gov/region2/qa/qa_documents/SOP%20HWSS-31.pdf
  17. United States Environmental Protection Agency. 2010. USEPA Contract Laboratory Program National Functional Guidelines for Inorganic Superfund Data Review (USEPA 540-R-10-011). Available from: http://www.epa.gov/superfund/programs/clp/download/ism/ism1nfg.pdf
Related Posts Plugin for WordPress, Blogger...

Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

Add Your Comments, Questions, or Feedback: